Case Report

Reperfusion Seizures: A Manifestation of Cerebral
Reperfusion Injury After Administration of Recombinant
Tissue Plasminogen Activator for Acute Ischemic Stroke
Faizan Hafeez, MD,* Mohammad A. Razzaq, MD,† Ross L. Levine, MD,† and
Mark Allan N. Ramirez, MD*

Reperfusion injury has been well described in medical literature; cerebral reperfusion injury is commonly seen in association with vascular surgical procedures such
as carotid endarterectomies and stent placement procedures. Cerebral reperfusion
injury can manifest as blood-brain barrier breakdown, cortical irritability, and
epileptic seizures. Seizures induced by cerebral reperfusion have not been documented or reported after thrombolytic therapy for acute ischemic stroke. We report
a patient who received intravenous recombinant tissue plasminogen activator
within 3 hours of stroke symptom onset and developed the new-onset symptom of
continuous, primary motor seizure activity within 20 minutes of recombinant tissue
plasminogen activator administration. These epileptic seizures originated in the
same area as the acute brain ischemia and occurred during the anticipated period
of cerebral reperfusion. In this article we describe a case report and then discuss the
pathophysiology and mechanisms that may underlie reperfusion epileptic seizures
as a manifestation of cerebral reperfusion injury. Key Words: Arrhythmia—
cerebral ischemia— epilepsy—plasminogen activators—reperfusion—thrombolytic
therapy.
© 2007 by National Stroke Association

In patients with acute ischemic stroke, timely intervention with thrombolytic agents can potentially induce arterial recanalization, reduce infarct size, and lead to clinical improvement or even complete recovery.
However, recombinant tissue plasminogen activator
(rtPA) administration is a double-edged sword and not
From the *University of Toledo Stroke Program, Department of
Neurology, University of Toledo Medical Center, Ohio; and †University of Wisconsin Stroke Program, University of Wisconsin Hospitals and Medical School, Madison.
Received August 4, 2006; revision received July 19, 2007; accepted
July 26, 2007.
Address correspondence to Faizan Hafeez, MD, University of
Toledo Medical Center, Department of Neurology, 3120 Glendale
Ave, Toledo, OH 43614.
1052-3057/$—see front matter
© 2007 by National Stroke Association
doi:10.1016/j.jstrokecerebrovasdis.2007.07.007

without potential harmful effects, which may include
hemorrhagic complications, direct toxic effects of rtPA on
neuronal tissue, and reperfusion-related injury. The toxic
effects of tissue plasminogen activator could result directly from enhanced neurotransmission or from immune
mobilization, or be triggered by its downstream effector
protein, plasmin.1 In this case report, we will focus on the
reperfusion-related injury.
Reperfusion injury can be divided into two categories:
hyperperfusion and normoperfusion. The first category,
hyperperfusion injury, is well described in medical literature and is commonly seen in association with vascular
surgical procedures such as carotid endarterectomies and
stent placement procedures. It can manifest as irritability,
headache, blood-brain barrier breakdown, and seizures.
However, the second category, which is normoperfusion
injury, has not been well established and there has been

Journal of Stroke and Cerebrovascular Diseases, Vol. 16, No. 6 (November-December), 2007: pp 273-277

273

F. HAFEEZ ET AL.

274

no report of epileptic seizures as a result of cerebral
reperfusion after thrombolytic therapy for acute ischemic
stroke.
We report a patient who received intravenous rtPA
within 3 hours of stroke symptom onset and developed
the new-onset symptom of continuous, primary motor
seizure activity within 20 minutes of rtPA administration.
These epileptic seizures originated in the same area as the
acute brain ischemia and occurred during the anticipated
period of cerebral reperfusion.

Case Report
Sudden stroke symptoms developed in a 78-year-old
man who experienced sudden onset of dysphasia and
right hemibody weakness. Stroke risk factors included
his age, sex, arterial hypertension, and hypercholesteremia. The patient had no history of transient ischemic
attacks, stroke-related problems, epileptic seizures, cardiac syncope, or neurologic or cardiologic symptomatology.
The physical examination was normal except for his
neurologic findings, which initially included global dysphasia, lethargy, left gaze preference, and weakness of
the right face, arm, and leg. The National Institutes of
Health Stroke Scale (NIHSS) score was 16. Relevant laboratory data produced normal findings and an emergent
noncontrast cranial computerized tomographic head scan
revealed entirely negative results. Magnetic resonance

Figure 1.
sphere.

imaging/diffusion-weighted imaging was not available
at the time.
Intravenous rtPA was administered in a dose and
method according to accepted protocols. Near the completion of this infusion, the patient became much more
alert and his comprehension was moderately improved.
The NIHSS score was 12 when he was transferred to an
intensive care department.
Approximately 20 minutes after this infusion was completed, he developed tachycardia, moderate hypertension, unresponsiveness to sensory stimuli, head turning
with forced gaze deviation to the right, and intense clonic
motor activity of the right arm. Intravenous administration of 2 mg of lorazepam resulted in cessation of seizure
activity, return of gaze to the neutral position, and improved alertness. A second noncontrast cranial computerized tomographic head scan revealed, again, negative
results.
Approximately 40 minutes after the initial seizure, the
patient experienced recurrence of identical seizure activity, which once again resolved with 2 mg of intravenous
lorazepam. A stat bedside electroencephalogram (EEG)
revealed sharp wave complexes in left parietotemporal
leads with superimposed slow wave activity throughout
the entire left cerebral hemisphere (Fig 1). The patient
then received a loading dose of fos-phenytoin. Relevant
laboratory data again revealed normal findings and there
was no recurrence of seizure activity. Immediate follow-up EEG after fos-phenytoin infusion revealed slow

EEG revealed sharp wave complexes in left parietotemporal leads with superimposed slow wave activity throughout entire left cerebral hemi-

REPERFUSION SEIZURES

275

Figure 2. Immediate follow-up EEG after fos-phenytoin infusion revealed slow wave activity in left parietotemporal leads and no epileptic activity.

wave activity in the left parietotemporal leads and no
epileptic activity (Fig 2). The patient was anticoagulated
with intravenous heparin and was maintained on phenytoin.
An EEG at 24 hours after stroke symptom onset revealed slow wave activity throughout the left cerebral
hemisphere but with no further sharp or epileptiform
activity. Magnetic resonance imaging on poststroke day 1
failed to reveal areas of brain infarction or hemorrhage
and magnetic resonance angiography revealed normal
extracranial and intracranial vessels. Transesophageal
echocardiography showed a freely mobile cord/strand
arising from the ventricular surface of his mitral valve
leaflet, thought to represent either a ruptured chordae
tendinea or valvular vegetation, considered a potential
source of emboli and the underlying origin for this patient’s stroke syndrome. Residual neurologic deficits
were limited to language and communication deficits,
with a NIHSS score of 1 on poststroke day 4. He was
discharged from our care at that point.

Discussion
In the current case, two acute yet separate neurologic
events occurred, both clinically separate entities, but temporally related to the same pathologic process. The first
event, a “negative” phenomenon, included sudden onset
of speech difficulty, gaze preference to the left, and right
hemibody weakness. This was an acute ischemic stroke in
the distribution of the left middle cerebral artery and was
cardioembolic in nature. The second event, a “positive”

phenomenon, included unresponsiveness, head turning
and gaze deviation to the right, and clonic motor activity
of the right arm. This was an epileptic event, associated
with an irritative EEG, and was believed to represent an
acute irritant focus of an ischemic-yet-hyperexcited collection of neurons (“umbra/penumbra”), undergoing
biochemical and electrophysiologic changes.
We believe that this epileptic event should be considered as a reperfusion-induced injury that is under the
normoperfusion category. This is quite different from
previously reported data of seizures caused by hyperperfusion after carotid endarterectomies.
In the cardiology literature, reperfusion arrhythmias
have been well documented after both administrations of
thrombolytic agents and during coronary revascularization procedures. Although several theories have been
proposed to explain the reperfusion arrhythmia phenomenon, the two leading hypotheses that explain the cellular
events involved in reperfusion irritability include: (1) the
oxygen free radical hypothesis2-5; and (2) the abnormalities of calcium hypothesis.6-8
Support for a free radical hypothesis was derived from
studies that showed that the incidence of reperfusioninduced ventricular tachycardia or fibrillation in rat
hearts could either be reduced by administration of antioxidant agents or increased by free radical– generating
systems.9,10 Garlick et al11 provided more direct evidence
for a measurable burst of free radical production at the
time of reperfusion. Further support for a free radical
hypothesis was derived from electrophysiologic studies
in which changes in action potentials, both early and late

F. HAFEEZ ET AL.

276

after depolarization, as well as increased automaticity of
cardiac tissues were induced by oxygen free radicals,
whereas these changes were attenuated by administration of antioxidants.12,13
Oxygen radicals have been known to cause peroxidation of polyunsaturated fatty acids in cellular membranes, thereby modifying the activity of a number of
translocating proteins in the sarcoplasmic reticulum and
the sarcolemma. This, in turn, caused an exaggerated
disturbance in calcium homeostasis and resulted in further calcium overloading in the cytosol after reperfusion.
This further resulted in excessive cycling, after depolarizations, and arrhythmogenesis.14
Support for a calcium hypothesis was derived from
observations that, as ischemic mitochondria start to regain their function after reperfusion, they expend their
energy stores on the massive uptake of calcium from the
cytosol that occurs with reperfusion and this severely
damages the respiratory chain and further decreases cellular energy production.6-8
The mechanism of reperfusion-induced irritability
after cerebral ischemia has not been fully elucidated,
either. It has been observed that reactive oxygen species, generated during the reperfusion period, may
trigger reperfusion injury. Mitochondrial calcium overload and a permeability transition of the inner mitochondrial membrane have also been shown to play an
important role in the production of reactive oxygen species and subsequent reperfusion injury.15 In contrast, free
radical scavenging enzymes have been shown in animal
models to be effective in ameliorating various types of
brain injury resulting from temporary brain ischemia and
reperfusion.16,17
Other mechanisms that have been implicated in the
pathogenesis of cerebral reperfusion injury include:
(1) depletion of the cellular antioxidant glutathione, leading to the oxidative conversion of protein thiols to disulfides, resulting in the loss of critical antioxidant enzymes
that had contained thiol-groups; (2) disturbances in the
receptor mediated poly-phosphoinositide signaling pathway; and (3) expression of adhesion molecules, such as
intercellular adhesion molecule-1, leading to production
of reactive oxygen species by polymorphonuclear leukocytes, suggesting the involvement of inflammatory and
immunologic reactions in reperfusion injury.15,18,19 The
role of nitric oxide synthesis and the role of excitotoxin
accumulation of glutamate have both been implicated
during the latter phases of reperfusion injury as well. It
has also been observed, in animals, that the excitotoxic
index (concentration glutamate multiplied by concentration glycine, product divided by concentration of gamma
amino butyric acid) decreased during periods of ongoing
cerebral ischemia and progressively increased on reperfusion.20 This excitotoxic index peaked at 32 ⫾ 2 hours
after onset of ischemia, was associated with nitric oxide
synthesis and accumulation of excitotoxins, paralleled

ischemic edema formation, and was associated with cellular irritability and frank seizure activity.20
It remains probable that during a period of cerebral
reperfusion multiple factors come into play, all subjecting
the ischemic neurons to a spectrum of biochemical and
electrophysiologic aberrations. During reperfusion, the
ischemic neurons may be prone to spontaneous electrical
activity, because of excessive excitotoxin accumulation,
reactive oxygen species production, calcium overload, or
a combination of these and yet undefined factors.
Epileptic seizure activity, as a potential functional
marker of cerebral reperfusion injury after thrombolytic
therapy for acute ischemic stroke, may occur through
similar mechanisms that account for the reperfusion arrhythmias in certain acute cardiologic treatments and
conditions. Prospective experimental and clinical trials
are needed for the validation of the clinical entity of
“reperfusion-induced epileptic seizures” and to further
define the apparent association of epileptic seizures induced by reperfusion after acute stroke interventional
treatment.

References
1. Sheehan J, Tsirka S. Fibrin-modifying serine protease
thrombin, tPA, and plasmin in ischemic stroke: A review. Glia 2005;50:340-350.
2. Jeroudi M, Harley C, Bolli R. Myocardial reperfusion
injury: Role of oxygen radicals and potential therapy
with antioxidants. Am J Cardiol 1994;73:2B-7B.
3. Bolli R, Patel B. Factors that determine the occurrence of
reperfusion arrhythmias. Am Heart J 1988;115:120.
4. Joshua I, Goldhaber W, Weiss J. Oxygen free radicals and
cardiac reperfusion abnormalities. Hypertension 1992;
20:118-127.
5. Bernier M, Hearse DJ, Manning AS. Reperfusion-induced arrhythmias and oxygen derived free radicals.
Circ Res 1986;58:331-340.
6. Nayler GW, Panagiotopuolos LZ, Daly MJ. Calcium mediated damage during post-ischemic reperfusion. J Mol
Cell Cardiol 1988;20:41-54 (suppl).
7. Bassett AL, Chakko S, Epstein M. Are calcium antagonists proarrhythmic? J Hypertension 1997;15:915-923.
8. Burashnikov A, Medvinski I, Hkin NI. The defining role
of calcium entering cells during ischemia as the mechanism for initiating reperfusion arrhythmias [in Russian].
Biofizika 1997;42:508-515.
9. Woodward B, Zakaria MNM. Effect of some free radical
scavengers on reperfusion induced arrhythmias in the
isolated rat heart. J Mol Cell Cardiol 1985;17:485-493.
10. Hearse DJ, Kusama Y, Bernier M. Rapid electrophysiological changes leading to arrhythmias in the aerobic rat
heart: Photosensitization studies with rose bengal-derived reactive oxygen intermediates. Circ Res 1989;65:
146-153.
11. Garlick PB, Davies MJ, Hearse DJ, et al. Direct detection
of free radical in the reperfused heart using electron spin
resonance spectroscopy. Circ Res 1987;61:757-760.
12. Pallandi RT, Perry MA, Campbell TJ. Proarrhythmic effects
of an oxygen derived free radical generating system on
action potentials recorded from guinea pig ventricular

REPERFUSION SEIZURES
myocardium: A possible cause of reperfusion-induced
arrhythmias. Circ Res 1987;61:50-54.
13. Barrington PL, Meier CF, Weglicki WB. Abnormal electrical activity induced by free radical generating systems
in isolated cardiocytes. J Mol Cell Cardiol 1988;20:11631178.
14. Hearse TJ, Tosaki A. Free radicals and calcium: Simultaneous interacting triggers as determinants of vulnerability to reperfusion-induced arrhythmias in the rat heart.
J Mol Cell Cardiol 1988;20:213-223.
15. Kuroda S, Siesjo BK. Reperfusion damage following focal ischemia: Pathophysiology and therapeutic windows.
Clin Neurosci 1997;4:199-212.
16. Francis JW, Ren J, Warren L, et al. Post ischemic infusion
of Cu/Zn superoxide dismutase or SOD:Tet 451 reduces

277
cerebral infarction following ischemia/reperfusion in
rats. Exp Neurol 1997;146:435-443.
17. Toyoda T, Kassell NF, Lee KS. Attenuation of ischemiareperfusion injury in the rat neocortex by the hydroxy radical
scavenger nicaraven. Neurosurgery 1997;40:372-377.
18. Panigrahi M, Sadguna Y, Shivkumar BR, et al. Alphalipoic acid protects against reperfusion injury following
cerebral ischemia in rats. Brain 1996;717:184-188.
19. Sun GY, Hsu CY. Poly-phosphoinositide-mediated
messengers in focal cerebral ischemia and reperfusion.
J Lipid Mediators Cell Signal 1996;14:137-145.
20. Tan WK, Williams CE, During MJ, et al. Accumulation of
cytotoxins during the development of seizures and
edema after hypoxic-ishemic injury in late gestation fetal
sheep. Pediatr Res 1996;39:791-797.